Face and Digit Detection with MNIST Dataset

This GitHub repository contains a Python project that uses the MNIST dataset for both face and digit detection. We employ three different classifiers: k-Nearest Neighbors (k-NN), Naive Bayes, and Perceptron, to recognize faces and digits within the dataset.

Project Overview

In this project, we explore the use of machine learning classifiers to distinguish between two types of images:

1. Faces: We aim to classify images as either containing a face or not.
2. Digits: We intend to classify images as digits and identify the digit represented.

We employ the popular MNIST dataset, which is widely used for digit recognition tasks, and a subset of the dataset containing face images.

Classifiers

We use the following classifiers to tackle the face and digit detection tasks:

1. k-Nearest Neighbors (k-NN): A simple yet effective instance-based learning method.
2. Naive Bayes: A probabilistic classifier based on the Bayes' theorem.
3. Perceptron: A single-layer neural network for linear classification.

Dataset

The MNIST dataset consists of a vast collection of 28x28 grayscale images of handwritten digits (0-9) and a subset of images with faces. You can download the dataset from the official MNIST website or use popular machine learning libraries like scikit-learn to access it.